Current techniques typically limit synthetic DNA to chain lengths of less than 150 bases and to modifications which can survive strong acid, base, and oxidants. The principal limitations are chemical side-reactions, in particular depurination. An enzymatic approach will allow the synthesis of long oligonucleotides containing chemically sensitive labels. This process is automatable and the unused reagents can be readily recycled. Phase I is designed to determine the feasibility of using polymerases to perform the automated synthesis of defined sequence oligonucleotides. The principal technical innovations in this phase of the proposal involve; 1) chemical synthesis of the nucleotide monomers; 2) synthesis of an appropriate solid phase resin; 3) modification of a DNA synthesizer to perform these syntheses; and 4) enzymatic synthesis of defined sequence oligonucleotides. In Phase II, the synthesis conditions will be refined to allow construction of very long chain oligonucleotides. Nucleotides containing backbone or modified bases will also be synthesized to make oligonucleotides containing chemically sensitive functionalities that are difficult to obtain using conventional means. The first of these products will be marketed during Phase II. PROPOSED COMMERCIAL APPLICATION: NOT AVAILABLE